Graphic system comprising a pipelined graphic engine, pipelining method and computer program product
    1.
    发明公开
    Graphic system comprising a pipelined graphic engine, pipelining method and computer program product 有权
    Graphisches System mit einer Graphikdatenpipeline,Verfahren zur Pipeline-Verarbeitung und Computerprogrammprodukt

    公开(公告)号:EP1496704A1

    公开(公告)日:2005-01-12

    申请号:EP03015269.8

    申请日:2003-07-07

    Abstract: A graphic system comprising a pipelined tridimensional graphic engine for generating image frames for a display inlcudes a graphic engine (110;210) comprising at least one geometric processing elaboration stages (111, 112), performing motion motion extraction. The engine also includes a rendering stage (113) generating full image frames (KF) at a first frame rate (F2) to be displayed at a second frame rate (F1), higher than the first frame rate (F2). The pipelined graphic engine further comprises a motion encoder (214) receiving motion vector information (MB) and suitable for coding the motion information e.g. with a variable length code, while generating a signal (R4) representative of interpolated frames (IF). The motion encoder (214) exploits the motion information (MB) as generated by the geometric elaboration stages (211, 212). A motion compensation stage (237) is provided fed with the signal representative of interpolated frames (IF) and full image frames for generating said the interpolated frames (IF). A preferred application is in graphic engines intended to operate in association with smart displays through a wireless connection, i.e. in mobile phones.

    Abstract translation: 一种图形系统,包括用于生成用于显示器的图像帧的流水线三维图形引擎,包括包括至少一个几何处理制作阶段(111,112)的图形引擎(110; 210),执行运动运动提取。 引擎还包括以比第一帧速率(F2)更高的以第二帧速率(F1)显示的以第一帧速率(F2)产生全图像帧(KF)的渲染级(113)。 流水线图形引擎还包括运动编码器(214),其接收运动矢量信息(MB)并且适于对运动信息进行编码。 同时产生表示内插帧(IF)的信号(R4)。 运动编码器(214)利用由几何制作阶段(211,212)产生的运动信息(MB)。 提供运动补偿级(237),其馈送代表内插帧(IF)的信号和用于产生所述内插帧(IF)的全图像帧。 优选的应用是旨在通过无线连接(即在移动电话中)与智能显示器相关联地操作的图形引擎。

    Graphic antialiasing method and graphic system employing the method
    2.
    发明公开
    Graphic antialiasing method and graphic system employing the method 有权
    Grafikantialisingverfahren und Grafiksystem,das dieses verwendet

    公开(公告)号:EP2026283A2

    公开(公告)日:2009-02-18

    申请号:EP08161219.4

    申请日:2008-07-25

    CPC classification number: G06T11/001 G06T11/40 G06T15/005

    Abstract: An antialiasing method, comprising:
    providing a first fragment;
    computing a first coverage area representing a portion of the first fragment covered by a first primitive;
    providing a second fragment juxtaposed to the first fragment and at least partially covered by a second primitive;
    processing the first coverage area to obtain a corrected coverage area indicative of a visible first fragment portion resulting from the juxtaposition of said fragments;
    applying an antialiasing procedure based on said corrected coverage area.

    Abstract translation: 一种抗混叠方法,包括:提供第一片段; 计算表示由第一原语所覆盖的第一片段的一部分的第一覆盖区域; 提供与所述第一片段并置并且至少部分地被第二原语覆盖的第二片段; 处理第一覆盖区域以获得指示由所述片段并置产生的可见第一片段部分的校正覆盖区域; 基于所述修正的覆盖区域应用抗混叠程序。

    Method for compressing high repetitivity data, in particular data used in memory device testing

    公开(公告)号:EP1416641A1

    公开(公告)日:2004-05-06

    申请号:EP02425660.4

    申请日:2002-10-30

    CPC classification number: G11C29/40 G11C29/44 H03M7/46 H03M7/48

    Abstract: Described herein is a method for compressing a sequence of repetitive data, which uses in combination one or more words with a format for non-compressible data and one or more words with a format for compressible data, in which a word with a format for non-compressible data is made up of a set of bits, in which the most significant bit is set at the logic value "1" and the remaining bits are the bits of a non-compressible datum to be encoded, whilst a word with a format for compressible data is made up of a set of bits, in which the most significant bit is set at a the logic value "0", the next five most significant bits indicate the total number of subsequent words which encode the sequence of repetitive data, and the remaining eleven bits indicate the number of times that the words indicated by the preceding five most significant bits are repeated.

    Abstract translation: 使用非可压缩和可压缩数据的组合对重复数据的序列进行编码。 非压缩数据的最高有效位(MSB)设置为逻辑1',而其余位是数据位。 压缩数据的MSB设置为逻辑值0'。 接下来的五个MSB显示编码重复数据所需的单词总数,而剩余的位表示重复次数。

    Built-in self test circuit for integrated circuits
    4.
    发明公开
    Built-in self test circuit for integrated circuits 有权
    Eingebaute Selbsttestschaltungfürintegrierte Schaltungen

    公开(公告)号:EP1388788A1

    公开(公告)日:2004-02-11

    申请号:EP02425519.2

    申请日:2002-08-08

    CPC classification number: G06F11/27 G01R31/3187

    Abstract: A built-in self-test (BIST) circuit adapted to be embedded in an integrated circuit ( 101 ) for testing the integrated circuit, including in particular a collection of addressable elements, for example a semiconductor memory. The BIST circuit comprises a general-purpose data processor ( 105 ) programmable for executing a test program for testing the integrated circuit. The BIST circuit comprises an accelerator circuit ( 113 ) cooperating with the general-purpose data processor for autonomously conducting operations on the integrated circuit according to the test program. The accelerator circuit comprises configuration means ( 201,203,233,301,303,501 ) adapted to be loaded with configuration parameters for adapting the accelerator circuit to the specific type of integrated circuit and the specific type of test program. ( FIG. 1 )

    Abstract translation: 内置的自检(BIST)电路,适于嵌入集成电路(101)中,用于测试集成电路,特别包括可寻址元件的集合,例如半导体存储器。 BIST电路包括可编程用于执行用于测试集成电路的测试程序的通用数据处理器(105)。 BIST电路包括与通用数据处理器协作的加速器电路(113),用于根据测试程序自动地对集成电路进行操作。 加速器电路包括适于加载用于使加速器电路适应特定类型的集成电路的配置参数和特定类型的测试程序的配置装置(201,203,233,301,303,501)。 (图1)

    Method and apparatus for texture compression
    8.
    发明公开
    Method and apparatus for texture compression 有权
    Verfahren und Vorrichtung zum Komprimierung von Texturen

    公开(公告)号:EP1443461A1

    公开(公告)日:2004-08-04

    申请号:EP03002205.7

    申请日:2003-01-31

    CPC classification number: G06T9/005

    Abstract: A method for texture compressing images having a plurality of color components (R, G, B) includes the step of defining color representatives for use in encoding by defining groups of colors for each said color component (R,G,B), and selecting for each said group a representative color for the group, the median color being chosen as the representative color of the group. Each said group is preferably composed of 3 to 15 colors and the method includes the step of computing, for each group, an error between each member of the group and said representative color of the group. Typically, the error is computed as the sum of the absolute differences (SAD) between each member of the group and said representative color of the group. In order to select each said group and then extract therefrom said representative color, a criterium is used selected from the group consisting of:

    selecting the group that minimizes said error by assuming each group comprised of the lower colors sorted in ascending order, wherein the same applies for the groups comprised of the higher colors,
    accruing the error as computed separately for two groups in all possible combinations and finding the minimum of the composite error.

    Abstract translation: 用于纹理压缩具有多个颜色分量(R,G,B)的图像的方法包括通过为每个所述颜色分量(R,G,B)定义颜色组来定义用于编码的颜色代表的步骤,以及选择 对于每个所述组,该组的代表颜色,中间颜色被选择为该组的代表颜色。 每个所述组优选地由3至15种颜色组成,并且该方法包括以下步骤:针对每个组计算组的每个成员与组的所述代表颜色之间的误差。 通常,误差被计算为组的每个成员与组的所述代表颜色之间的绝对差(SAD)之和。 为了选择每个所述组,然后从其中提取所述代表性颜色,使用选自以下的标准:选择通过假设由按升序排列的较低颜色组成的每个组,使所述误差最小化的组,其中相同 适用于由较高颜色组成的组,在所有可能的组合中分别计算出两组的误差,并找出复合误差的最小值。

Patent Agency Ranking